Street-Level Organizations and the "Real World" of Workfare: Lessons from the US
Evelyn Z. Brodkin, Associate Professor in the School of Social Service Administration, University of Chicago
By the end of the 20th C., workfare and labor market activation policies had become increasingly common around the globe. These developments provoked debate over whether these policies were advancing a project of commodification (Esping-Andersen 1990) by marketizing citizenship or a project of inclusion by bringing marginalized groups into the economy and society or whether they were advancing inclusion by supporting labor market participation.
Assessment of this developments is complicated, in part, by the bewildering array of policy labels under which they have occurred, among them welfare-to-work, labor market activation, jobseekers' allowance, Hartz reforms, and revenu minimum d’insertion. Yet, the many variants of workfare and activation tend to draw on a common assortment of features, among them, potentially enabling provisions for training and work. The shift toward conditionality in benefits is part of a broad trend that, in some countries, has eroded or even replaced rights-based benefits that were previously conditioned primarily on need, legal right (e.g., compensation to unemployed workers), or family status (e.g, welfare benefits for families with children).
These policy complexities are further compounded by the uncertainties of implementation. Exactly what goes on under these policy labels is far from obvious. In this context, efforts to characterize policies on the basis of their formal provisions or legislative language alone are apt to reveal more about what policy says than what it does. On the other hand, efforts to assess what works by using benchmarks or other selected outcomes provide important insights, but also have their limitations. Specifically, they tend to rely on measures that essentially assume rather than investigate what happens on the ground in the name of policy, that is, how policy works.
As a general matter, the practices that occur under the rubric of policy are too often assumed rather than interrogated in policy studies, perhaps, in part, because street-level practices are difficult to examine, largely operating beyond the language of formal policy, metrics of administrative data, and outcomes measures. However, the significance of these practices is evident. Simply put, decades of implementation and organizational research make clear that what you see in terms of formal or written policy may not be what you get.
In this article, I make the case for a street-level approach to research, not to replace other approaches, but to address the gaps they leave between analysis of formal policy provisions and measurable outcomes. These gaps may be regarded as the missing middle of policy analysis. A Street-level approach provides a strategy for directly interrogating how policy works in practice rather than imputing meaning to formal law, administrative categories, or policy rhetoric. This approach extends beyond the boundaries of conventional implementation analysis, in part, because it requires the analyst to ask "what else" happens in under rubric of policy beyond what policymakers claim or administrative metrics reveal. The approach I discuss is theory-driven and systematic, thus different from the process studies that often accompany policy evaluations and from other descriptive forms of implementation research.
This approach provides an empirical strategy for directly examining the practices of workfare in order to make visible and assess the mechanisms that constitute the processes of commodification or inclusion on the ground. I look closely at the case of workfare, often called "welfare reform," in the United States.
Workfare in the United States
As a legislative project, welfare reform was highly contentious, some 25 years in the making before culminating in the enactment of the Temporary Assistance to Needy Families (TANF) Act in 1996. This Act marked an end to the New Deal innovation of a federal entitlement to assistance for poor families with children. It reflected a turn in social policymaking, one in which income support would be routinely conditioned on work or (in the arcane language of welfare policy) "work activities." It also meant that welfare could not be relied on as an alternative to market-derived income. The federal government would finance TANF income benefits for only a limited duration, largely (although not entirely) without regard to family poverty or circumstances in the larger economy that might create economic need for longer periods.
Still, the Act was not completely one-sided. As is often the case in social policymaking, it combined regulatory features with enabling features that offered supports to those seeking to make it in the labor market. TANF and companion welfare-to-work legislation provided funds to states for education and training, child care, transportation subsidies, and other supports manifestly designed to promote and facilitate work. It also allowed extended benefits for a limited number of individuals determined to have “barriers” to work, for example, health needs and caretaking responsibilities.
The law established a framework that policymakers holding different views of welfare reform could find acceptable (Weaver, 2000). It contained provisions that offered disadvantaged adults help entering and staying in the labor market and provided income support during periods of unemployment. In this respect, welfare reform had features associated with a policy of inclusion. This term, although imprecise, refers both to policy features that provide access to a basic income and those that provide supports for disadvantaged and lower-wage workers. The welfare reform law also contained provisions explicitly limiting the duration of income assistance and making benefits conditional on work or "work effort." These provisions are consistent with a project of commodification.
Commodification is a concept that gained analytic importance through Gosta Esping-Andersen's (1990) Three Worlds of Welfare Capitalism, in which welfare states were grouped according to the extent to which they de-commodified citizens. The term indicted the degree to which states support individual well-being based on citizenship rather than on work effort, and, as a consequence, relieve the pressure to work at any cost. Esping-Andersen's analysis distinguished among social democratic, corporatist, and "liberal" welfare states (moving from most de-commodifying to least). The liberal regimes were those (like the U.S.) that offered the fewest alternatives to market-derived income. Without weighing in here on the subsequent debates this view inspired, the concept of commodification offers a heuristic for conceptualizing the regulatory features of welfare reform and juxtaposing them against those that appear to advance inclusion.
In terms of its formal policy provisions, welfare reform incorporated two rather different policy projects: one that had the attributes of a project of inclusion, the other a policy of commodification. Policymakers on either side of debate could present welfare reform as they preferred it or just blur the lines entirely as President Bill Clinton did when he famously vowed to "end welfare as we know it," inviting listeners to fill in the blanks as to what that meant.
However, the law's formal provisions were only a prelude. After a law is enacted, another stage of policymaking begins. In the next stage, policymaking is less visible and largely indirect, taking place through the street-level practices that mediate policy on the ground (Brodkin & Majmundar, 2010). Although formal policy sets the framework for these practices, it does not fully determine what they will be and how they will evolve. The processes of policy delivery allow space for discretion and, in a sense, require it in order to reconcile the contradictions and ambiguities that legislative bargaining and compromise create.
In the case of TANF, these contradictions were expressed, not only in the more obvious policy features that have received the greatest attention (e.g., work requirements and work supports), but also in provisions that made various demands on street-level organizations implementing the law. For example, policy required state welfare agencies to assess and respond to individual needs and capacities and to provide work supports, activities suggestive of inclusion. Other provisions required state welfare agencies to enforce work and subject claimants to an increasing variety of categorical and procedural tests and behavioral requirements, activities suggestive of commodification (Brodkin & Majmundar, 2010; Diller, 2000).
Presumably, one could review all of TANF’s formal provisions and assign them to categories. But this would be a futile exercise, because the provisions alone do not indicate how they really worked. In the past, work requirements were often ignored or eased in day-to-day agency practice (Handler & Hasenfeld, 1991). Similarly, income benefits provided by law were not necessarily accessible in practice (Brodkin & Majmundar, 2010; Handler & Hollingworth, 1971; Piven & Cloward, 1971). Simply put, inscribing either rights or obligations in formal policy does not make it so.
In order to assess whether welfare reform - and the variant of workfare it created - constituted a project of inclusion or commodification (or something else), analysis of formal policy provisions is insufficient. It is necessary to look more closely – beyond formal provisions and stated intentions - to determine how the real world of workfare was constituted in street-level practice.
The empirical challenge is to illuminate how workfare was constructed in the everyday organizational life of a major urban welfare bureaucracy. The approach offered here is markedly different from much of the familiar policy research that either studies welfare-to-work as a policy to be understood either on its own terms (that is, by measuring changes in caseloads, costs, work hours) or in terms of individuals (for example, by looking at the behavior or lives of poor women and their families). Instead, this article offers an organizational view, one that does not presume what policy is, but rather looks at what street-level organizations do in order to illuminate policy-as-produced and explore its implications. This inquiry asks: What were the practices of workfare; what shaped their evolution; and how did street-level organizations construct inclusion, commodification, or possibly something else on the ground?
Street-Level Organizations and Policy Research
This article adopts an analytic framework that places street-level organizations (SLOS) at the forefront of analysis in order to directly examine the "missing middle," the organizational practices through which policy and politics are mediated (Brodkin 2013b). SLOS are situated at the very center of the missing middle where they, in effect, form the operational core of the welfare state. By SLOs, I specifically mean those agencies and governmental departments that directly deliver policy to people.
A subset of these organizations is the widely familiar street-level bureaucracies (SLBs), which came to occupy analytic importance through the seminal work of Michael Lipsky (1980). His insightful treatment of the large public bureaucracies that, at one time, were the main providers of human services and social policy has generated a rich empirical literature aimed at understanding the practices of government agencies operating at the front lines of policy delivery. SLBs can be distinguished from other public bureaucracies by their role in direct provision and by the availability of discretion at the front lines of organizational practice. In SLB research, the main challenge is to understand how discretion is exercised in public agencies and what that means for the production of policy on the ground.
The study of SLOs builds on this line of inquiry, but takes account of the new organizational environment in which policy delivery occurs today. In the United States and many other countries, the policy world has changed; large public bureaucracies are not the only, or necessarily the most significant, locations in which the work of the welfare state takes place. Policy may be delivered through a variety of organizational forms - among them, public bureaucracies, nongovernmental organizations, for-profit firms, and mixed public-private arrangements. In addition, there have been significant developments in management and governance that are transforming the conditions under which policy delivery occurs. In many countries, strategies associated with new public management (NPM) - among them devolution, contracting, and performance management - have reshaped the conditions of street-level work.
These regimes of managerial control have more than technical interest. They may shape policy projects by indirectly and (largely) invisibly influencing the street-level policy production. Although these reforms generally are advocated as strategies for improving, efficiency, effectiveness, and/or accountability. But, from a street-level perspective, they derive their significance from how they shape who does what (e.g. reorganization, devolution, privatization) and by influencing how they do it (e.g., performance management and various kinds of financial incentives). The SLO approach takes analysis beyond the implicit political neutrality of management and governance strategies to investigate how they play out on the ground and whether, in practice, they skew policy in systematic -- and politically significant -- ways. This approach is part of a broader critical turn in public management research (Considine 2000 and 2005, Moynihan and Herd 2010, Radin 2006, and Van Slyke 2003).
It also builds on a fundamental recognition that people experience the welfare state within specific organizational settings, not as abstracted rules or ideals. Nor are these experiences simply random or largely idiosyncratic. Rather, they are structured by everyday organizational practices. If one wants to understand broad issues involving how welfare state institutions mediate between citizen and market, it is important to look directly and microscopically at the organizations in which this mediation actually takes place (Brodkin 2013b). It is within these politically-embedded organizations that the "real worlds of welfare capitalism" (Goodin et al 1999) and, potentially, the commodification of citizenship or economic inclusion assume their practical shape.
How Governance and Management Matter
With its mix of policy complexity and delegated discretion, TANF effectively imbued street-level organizations (SLOs) with a major role in shaping policy. To the extent that the project of reform depended on SLOs, it created a new problem: How to manage the work of policy implementation in a complex and discretionary environment? In this case, Congress used governance and management provisions to address – if indirectly – both the political problem of "what counts?" and the managerial problem of implementation.
How were these provisions used to manage policy work? TANF devolved considerable autonomy to states to set policy and organize practices. But mistrusting states to be tough enough on work, Congress set benchmarks for welfare-to-work participation, specifying a menu of acceptable activities and establishing an escalating participation rate. By the five year mark, 50 percent of welfare recipients in single-parent households were required to participate in work activities for 35 hours per week. Ninety percent of adults heading two-parent households were required to participate in work activities for 30 hours per week. The law also specified the kinds of activities that would count toward meeting these quotas, among them, paid work, job search, unpaid workfare (in which recipients "worked off" their welfare benefits at minimum wage or provided child care for other welfare recipients). It also limited the use of education and vocational training as countable activities to meet participation benchmarks.
In addition, TANF rewarded states for reducing their caseloads, in part, by providing a financial bonus to states that cut caseloads. The law disallowed credit for caseload cuts achieved by imposing new eligibility rules that excluded categories of poor adults. But otherwise, states benefited from cutting caseloads whether by improving the economic prospects of poor families or by pushing people into bad jobs or simply off welfare.
Arguably, these provisions were designed to encourage good programmatic practices and administrative prudence. However, when incentives are unbalanced, encouraging caseload reduction without penalizing or otherwise discouraging wrongful exclusion or poor quality services, organizational theory would predict that quality is likely to be compromised and that organizational practices will be skewed toward those dimensions of performance that "count." The relationship between performance incentives and practices, although indirect, can be pervasive and powerful.
From an organizational perspective, TANF's governance provisions can be viewed as effectively defining the policy goals that mattered, emphasizing accountability for work enforcement and caseload clearance. These provisions were as significant for what they did include as for what they did not. TANF's governance provisions did not hold states accountable for providing the right services to the right people, for assuring access to eligible families, for providing any particular quality of services, or for improving the lives of poor families. In a little-remarked nod to this last concern, there was brief debate in Congress over an amendment to measure TANF's anti-poverty effects on children. However, these concerns were not incorporated into the final bill. On the whole, governance provisions emphasized practices more closely associated with a project of commodification than one of inclusion.
This raises critical questions for analysis. How did street-level organizations adapt to new policy and governance arrangements? What did these adaptations mean for the real world of workfare and the production of inclusion or commodification on the ground?
Research Strategy in Brief
This inquiry builds on two well-known scholarly traditions. One is associated with organizational research that examines policy delivery at the street-level. The study of "street-level bureaucracy" investigates how policy is made through the discretionary behaviors of those at the front lines of policy delivery (Lipsky, 1980). A second tradition is associated with the Chicago school of sociology in that it grounds analysis of social relations in particular settings in order to give it depth and context. In this case, the analysis of policy processes is located in specific organizational settings in order to develop a deeply-contextualized understanding of how they are formed and what they do. This approach sees street-level organizations as politically-embedded, both reflecting and refracting the larger environment in which they operate.
Empirically, this analysis draws on my organizational ethnographic research in three selected Chicago welfare offices as well as findings from my collaborative quantitative studies of organizational practices (Brodkin & Majmundar, 2010). The Chicago case was selected, not to represent all welfare agencies, but to probe everyday life in a big city welfare agency, responsible for the vast majority of the state's welfare caseload and serving an urban population – a population whose poverty was arguably at the heart of welfare reform. Field research was conducted in three welfare offices purposefully selected to focus on locations serving areas of concentrated urban poverty, while varying in theoretically relevant organizational elements, including office size, race/ethnicity, rate of caseload decline in the first year after welfare reform, and neighbourhood. Intensive field research continued over a three-year period (1999-2001) with follow-up visits and interviews in the subsequent two years.
The research utilized methods of semi-structured and open-ended interviews, observation, and on occasion, participant-observation. This included shadowing caseworkers through their days and simply "hanging around" in order to observe routine organizational and case management activities. It also included extensive observations of caseworker-client interactions that were part of the processes of assessment and eligibility review, training sessions, job clubs, and management meetings. In addition, interviews were conducted with state-level managers, as well as the regional-level managers responsible for each of the local offices studied and the top 3 - 5 management staff in each office. Other data sources include agency and business documents, case materials from legal challenges and complaints, administrative data and reports, and evidence from evaluation and research studies.
The analysis uses the extended case method, in which "the significance of a case relates to what it tells us about the world in which it is embedded" (Burawoy 1998:281). The first-order objective of an organizational ethnography is to reconstruct agency practice in terms of its own internal logic. This involves a systematic examination of both the conditions of work and the content of practice, moving heuristically between the two in an effort to explain the particular form that implementation takes in specific settings. The second-order objective is to relate these practices to the larger political context in which they are embedded. In this case, the objective is to consider how street-level practices constructed workfare through their informal practices and whether those practices advanced a broader project of either inclusion or commodification.
The following discussion focuses on practices that had particular significance for shaping that world and creating commodification or inclusion in everyday organizational life.
Simplification, Reductionism, and Access to Services
As discussed, street-level practices associated with achieving participation rates and caseload reduction was closely monitored. This occurred on the local office level, as well as in smaller (4-7 person) units, where supervisors reviewed performance metrics regularly, commonly intensifying surveillance and pushing for improvement when caseworkers fell short. In this context, caseworkers used their discretion to devise adaptive strategies that reflected a street-level calculus of costs and constraints. These strategies effectively reduced the cost of complex tasks such as assessing, sorting, and supporting individuals, tasks that were ostensibly part of a policy design aimed at human capital development and labour market inclusion. One street-level adaptation involved a variety of reductionist strategies that lowered the cost of what is familiarly known as "meeting the numbers." These strategies included simplification, categorization, and redefinition of the casework task to enhance processing efficiency. These patterned adaptations did not need to be conscious or purposeful in order to have systematic effects, that is, positive effects on processing efficiency, as well as negative effects in terms of responsiveness to need and wrongful exclusion.
Reductionist adaptations enabled caseworkers to expedite processes requiring them to assess individual client needs and devise programmatic plans to fit them. For example, one caseworker saved time by preparing assessments in advance of meeting her clients. Another caseworker, a specialist in a referral unit for clients with drug and alcohol problems, demonstrated how she expedited the process of determining “barriers” to work and service referrals permitted an observer to sit in on the assessment process. With each client, she quickly went down the written checklist of questions – "do you have an abuse problem?" – ticking boxes on the list. She then, almost instantly, came up with a program referral. Asked how she could manage such a complicated task so quickly, she responded that selected programs on the basis of zip code. When asked whether other elements of fit entered into consideration, she said that she knew little about either the programs or the individuals, but assured the observer that it was best for clients to attend programs nearest to their homes.
Although specific strategies varied, speed-over-need adaptations generally functioned to maximize processing efficiency. The logic of such practices was illuminated in extended observations of one caseworker, the most celebrated in his office, recognized by managers for his exceptional productivity. This caseworker (who I will refer to as Mr. Frank) consistently held the top position in the "Fifty Percent Club," an honor conferred on caseworkers who exceeded performance standards by achieving work participation rates of 50 percent or higher. Observation provided an opportunity to ascertain how he achieved that success. In a sense, he was a master of strategies that enabled him to simplify complex job tasks (such as assessments and referrals), limit demands for individualized responsiveness, and suppress potential complaints.
First, Mr. Frank achieved scheduling efficiency by arranging all appointments for only two days of each week (leaving other days free for paperwork) and scheduling each of his clients (on average about 10 per day) for the same hour of 9 a.m. This strategy enabled him to shift the cost of waiting for appointments from him to his clients. Given a clientele composed largely of single mothers, there were those for whom waiting meant they might not be able to pick up their children at school. Those who did not remain were subject to penalties for noncompliance and termination of benefits. Mr. Frank, at times, advised clients to bring money for pay telephone to call him from the waiting room. But he was rarely available to answer the phone, and the office lacked a functioning message system.
In addition to cost-shifting, Mr. Frank also developed simplification strategies that enhanced efficiency in the assessment process. He discouraged claimants who tried to present complicated personal issues to him, in part, by asserting that their problems were outside his domain (regardless of whether that was the case). He used informal decision rules that saved time and allowed him to ration scarce services. For example, one week he referred nearly every client to a single welfare-to-work service provider. When asked why he made the same referral for each client, he replied that he sent them to his "favorite" program, his favorite being the one that submitted paperwork to him on time, enabling him to process casework more expeditiously.
Asked if that was the right match for each client, Mr. Frank readily acknowledged that he did not know. He explained that he regarded these placements as an informal test. Those who met this test proved they “deserved” an opportunity later to move into a more suitable program. Those who complained or did not comply, he regarded as undeserving. Essentially, Mr. Frank demonstrated the classic street-level strategy of redefining his clients to fit his processing routine. By this strategy, those demanding responsiveness to their individual situations were likely to be labelled as troublesome.
Mr. Frank readily acknowledged that he assigned individuals to service providers that he knew did not deliver on their promises. He described one agency, which told clients that after 40 days of unpaid work, they would be placed in paid positions. But "they're still in work-first for 6-9 months." When asked why he continued to use this agency, he explained, "That's a safety valve to lower your caseload." Compliance would raise his participation rate; non-compliance would reduce his caseload. Either contributed to his efforts to excel in measured performance, and he did. Mr. Frank added that the hardest thing about his job was cutting people's benefits when they did not comply with his decisions. He explained that "it was really hard at first. I was thinking about these poor children without food.... But it comes down to whose job do I like more - theirs or mine?"
No single example can capture the variety of street-level adaptations favoring speed over need. While these informal practices were functional for the welfare agency and state government in saving money and meeting benchmarks, they were dysfunctional for welfare recipients seeking to make it in the labor market or trying to maintain access to income benefits provided by law.
Job Club Rituals, Symbolism and Realities
Street-level work involves more than paper pushing or eligibility reviews. Caseworkers also are expected to function as labor market intermediaries and job search coaches. Although many states contract out these types of services, the first step toward either services or employment is participation in local office job clubs, where the welfare-to-work process begins.
Under TANF governance provisions, participation in job clubs counted toward federal work participation benchmarks. Federal law set participation hours and a required that participants search for jobs, but otherwise, did not specify what job clubs should do. In Chicago, operational discretion largely was delegated to job club staff. There were no formal criteria that qualified caseworkers for job club positions. From the point of view of caseworkers, these positions were desirable because they allowed for greater autonomy and less paperwork than standard casework positions.
Once assigned to these positions, what were the conditions under which job club practices took shape? First, resources were limited. Generally, leaders received a room to meet, access to the department's general database of employers, and, at times, a TV and video player. They did not receive special training in the provision of services for job seekers, yet had discretion in designing programmatic activities that would take place in job club sessions. Second, they were subject to management oversight focusing on compliance with attendance and job search requirements, both of which were necessary in order for job club participation to count toward the agency’s participation benchmarks.
Observation in scores of job club sessions suggested that, on the whole, caseworker practices under these circumstances were idiosyncratic, dependent on available resources and the skills and inclination of job club leaders. In a context in which staff did not have access to a steady supply of suitable jobs or a clear technology for supporting job-seeking efforts, they developed strategies that enabled them to manage their responsibilities, in part, by developing ritualized activities that filled the time, met performance benchmarks, and possibly offered affective benefits of hope and support for some participants. That said, these rituals could be strikingly disconnected from the realities of the lower-wage labor market that was the primary employment source for welfare recipients.
For example, at one job club session, a group of women, mostly minority and unskilled, were told to watch and discuss a film about searching for a job. The film featured young, white male professionals preparing to interview for white-collar business jobs. Job seekers were advised to buy a good wool suit and to have a stationer copy their resumes on high quality paper. While women (mostly African-American or Latino) in sweat suits watched and discussed the film in one room of the welfare office, in another room, employers from freight service companies were interviewing other recipients for packing and shipping jobs. One company recruiter described her hiring criteria this way: "I don't care if they wear a nose ring and have tattoos all over their body. If they can be at our building at 5 a.m. and lift 60 pounds, I'll hire them." However, the company found few recruits, in part, because the shipping center was in a suburb about one hour by car from the welfare office and unreachable by public transportation.
Ritualization could take many forms. In another welfare office, a job club leader fashioned herself into dispenser of self-help wisdom and inspirational personal stories. Participants could be seen emerging from her sessions pumped up and chanting positive slogans ("we can do it!"). Despite the obvious enthusiasm of the participants, it was not clear that the types of job club wisdom dispensed in these sessions, mostly about appearance and personal characteristics (sometimes referred to as "soft skills"), were necessarily relevant to their employment prospects. Job club staff did not receive any systematic feedback regarding the quality or effectiveness of their efforts. The only performance that mattered was that linked to benchmarks for client attendance and number of job applications submitted.
The disjuncture between the ritualized world of the job club and the real world of lower-wage work appears, at first glance, to illustrate a standard implementation problem, in which lower-level irrationality and discretion undermine higher-level objectives. However, this view ignores the terms of governance that make such practices rational for both the street-level worker and the organization itself. Given the limited resources under which job club staff operated and the skewed performance for which they were accountable, street-level made ritualization was an adaptive strategy. Arguably, this strategy was not only adaptive at the individual (caseworker) level, but also at the agency level (and possibly beyond) to the extent that it appeared to provide help to job seekers, in effect, deploying "myth and ceremony" to obscure the absence of material support (Hasenfeld, 2000).
In this sense, job clubs may be regarded as more symbolic than practical in their function, substituting ritual of uncertain value for practices more likely to contribute to human capital development. Symbolically, they appear to engage in efforts to advance inclusion, while, instead, reaffirming the problematic moral status of those receiving cash aid without work (Handler & Hasenfeld, 1991; Hasenfeld, 2000). To the extent that the practices of job clubs generally developed in a context of resource constraints and lack of accountability for value-added, one would have to conclude that their everyday practices functioned in ways more consistent with a project of commodification than one of inclusion.
De-legitimating Claims on the State: Teaching Casework
A project of commodification can be said to advance when policies and practices reduce the ability of citizens to make claims on the state for protection from market vulnerability and receive income support that allows, at minimum, for the basic necessities of life. Commodification may be produced through informal organizational practices that function to de-legitimate claimants and claimsmaking, in part, by redefining claimants from citizen-in-need (entitled to responsiveness from the state) to "dependent" (and potentially inauthentic) claimants, subject to caseworker discipline and sanctions.
De-legitimation may not be a matter of conscious strategy. Rather, it may be a by-product of the street-level adaptation to work demands in the context of limited resources and skewed performance incentives. In this case, as discussed, street-level practitioners informally adapted to organizational conditions, in part, by limiting access; they adjusted to resource limitations by using strategies such as simplification and ritualization. However, these strategies alone could not adequately suppress the demand side of workload equation. Moreover, they had the potential to create cognitive dissonance for caseworkers who may have wanted to do more to help, but found they could not, creating the potential for street-level resistance to managerial pressures. It is in this context that de-legitimation emerged as an additional adaptive strategy.
Usually this kind of subtle adaptation must be inferred indirectly from extended field research. But on rare occasion, it was possible to observe informal processes of de-legitimation taking shape more directly and actively encouraged by management, although certainly not portrayed or understood in these terms. Observation of local office sessions called "staffings" (a term borrowing from medical terminology for meetings in which medical professionals jointly review key cases) provided a window into this process. At staffings, welfare office managers and supervisors met with a professional consultant (a social work Ph.D.) to review selected cases, specifically targeting those that were not contributing to participation benchmarks.
At one staffing, the consultant presented the agency's formal casework model, called "The Pipeline," which she described as a "tool to determine client's needs." In a rather Orwellian spin, the pipeline was actually used as a tool to convert clients with problems into problem clients. It sorted them into three categories representing their position in the welfare-to-work process. They were either:
· "Lost" - "avoids contact with service providers, angry and resentful";
· "Stuck" - "follows only the letter of the plan, easily discouraged"; or
· "Flowing" - "is on time for appointments, enthusiastically participates."
In a rare formalization of client redefinition, the agency used the pipeline to explain why clients weren't working in terms of their own personal deficiencies in 'belief,' 'attitude,' or 'behavior,' not in terms of their skills, health, family issues, the availability of job opportunities, or the difficulties of managing work and family in the lower-wage labor market. Clients who were deemed at risk of imminent "failure" to make it through the process were informally referred to as "waste."
The view of clients as deviants to be brought into line was reinforced in staffings, where casework supervisors were called to account for clients who were not meeting work participation requirements. In virtually each case, supervisors who claimed their clients could not work, because they were homeless, mentally ill, or had some other serious problem, were pressed to reconsider these clients as suspect, at best, unwilling to comply, at worst. This included cases in which the formal record documented serious health or mental problems. When a supervisor attempted to argue that a client with a documented mental illness could not work, the staffing leader insisted: "If you can't do anything, that means you're in a vegetative state. There are no medical exemptions - only barriers."
Staffings were conducted in ways that instructed street-level supervisors and their caseworkers to redefine both the task and the client, a strategy that ordinarily occurs in less visible form. In effect, staffings defined the task of casework, not as one of assessing or responding to individual needs and circumstance, but as one of pushing clients through the "pipeline" and placing blame for failure largely on the clients themselves. At staffings, unit supervisors were called to account for cases that did not contribute to performance benchmarks. They were instructed to regard claims of individual need with suspicion and nonworking claimants with skepticism. At one session, the staffing leader passed around the file of a recipient who had received an exemption from work requirements because she was homeless (an exemption permitted by law). The leader challenged the group to look at the photo in the file. "Does she look homeless to you?" she demanded.
The sessions made clear that supervisors and their caseworkers could – and would – be called on the carpet for excusing individuals from work requirements, even if those excuses were arguably permitted – ostensibly required -- under good cause exemptions in the law. Supervisors brought this approach to their casework units, through means of instruction, surveillance, and pressure to justify any case that did not contribute to participation benchmarks. For caseworkers, the option to consider clients’ problems and offer exemptions became that much harder.
These disturbing examples are striking given that welfare law permits, even requires, exemptions in these cases. But these features of law were less relevant to street-level practice than governance provisions measuring work participation and caseload benchmarks. Of course, the patterns of street-level practices described here might have been disrupted if there were substantial counter-pressure by citizen/claimants that altered the choice calculus for caseworkers. However, opportunities to assert an alternative conception of citizenship rights have traditionally been quite limited in welfare settings. Far more relevant to everyday practice were general terms of TANF governance that were largely indifferent to dimensions of performance related to responsiveness to claimant needs or attentiveness to service quality.
Reconsidering the "Real World" of Workfare
Welfare reform and its workfare provisions were highly contested as a matter of explicit legislative policymaking. Making the case for welfare reform, proponents argued that it offered a strategy for bringing disadvantaged citizens into the market economy and out of marginalized status, in effect, a policy of inclusion. Opponents argued that workfare provisions would "regulate the poor" (Piven & Cloward, 1971) and either drive them into a labor market where they would face economic hardship and insecurity or simply drive them to destitution. From that perspective, welfare reform constituted an instrument of commodification. TANF's formal provisions could be marshaled in support of either view. They limited income benefits and imposed work requirements. But they also introduced new work supports and maintained assistance (if temporarily) for those judged to have special "barriers" that prevented employment.
This analysis brings an organizational perspective to this debate, directly examining the street-level practices through which the real world of workfare was constructed. The approach adopted here permits analysis to probe beneath categorical policy terms, managerial metrics, and even the ways in which street-level practitioners and beneficiaries perceive their experiences. This close examination of street-level practices in one big city agency and at one (extended) point-in-time may this limit generalization in certain respects. But it also reveals how common, structural features have shaped the practices of workfare and how street-level workfare practices developed at the intersection of formal policy provisions, organizational arrangements, and managerial instruments. One would expect different conditions to produce different practices. However, the common elements of workfare as a policy strategy and the extensive use of new managerial instruments as an administrative strategy are hardly unique to this case.
These arrangements indirectly, but systematically, fostered practices that emphasized welfare reform's commodifying features and undermined its supportive ones.
This study’s portrait of the real world of workfare is more consistent with a project of commodification than one of inclusion. It also reveals how the practices of commodification take shape in everyday organizational life. Formal policies set the stage; but informal practices seal the deal. This is especially problematic because informal practices are not readily visible or easy to discern. Nor are they captured in the limited language of administrative metrics.
From a rational choice perspective, it is logical for caseworkers to avoid administrative activities that are costly to them in time and effort (e.g. processing exemptions). It also is rational for caseworkers to avoid grappling with the complicated problems that beset some of those individuals facing sanctions. In some circumstances, it would require caseworkers to make fairly heroic efforts in order to take the more responsive course, making it all the more remarkable when they do. The arrangements described in this chapter largely provided the appearance of accountability while leaving welfare agencies virtually unaccountable for quality of service provision and for assuring needy families access to benefits the law provides. This unbalanced approach to welfare administration, while it did not directly assert a policy bias, indirectly skewed street-level practices, exacerbating workfare’s already considerable commodifying tendencies.
This analysis underscores that the transformation of the welfare state, and political projects of inclusion and commodification, cannot be adequately understood by examining formal policies alone or outcomes that are only loosely linked to them, although both are clearly important. It is also necessary to examine the organizational practices of the welfare state -- how street-level organizations construct policy in practice -- in order to more adequately expose, assess, and potentially reform the mechanisms that create the terms of inclusion and commodification on the ground.
Brodkin, E. Z. (2003). Street-level Research: Policy at the Front Lines. In T. Corbett & M. C. Lennon (Eds.), Policy into Action: Implementation Research and Welfare Reform. Washington, DC: Urban Institute Press.
Brodkin, E. Z. (2013a). Commodification, Inclusion, or What: Workfare in Everyday Organizational Life. In E. Z. a. M. Brodkin, Gregory (Ed.), Work and the Welfare State: Street-Level Organizations and Workfare Politics (pp. 143-166). Washington, DC: Georgetown University Press.
Brodkin, E. Z. (2013b). Work and the Welfare State. In E. Z. a. M. Brodkin, Gregory (Ed.), Work and the Welfare State: Street-Level Organizations and Workfare Politics (pp. 3-16). Washington, DC: Georgetown University Press.
Brodkin, E. Z., & Majmundar, M. (2010). Administrative Exclusion: Organizations and the Hidden Costs of Welfare Claiming. Journal of Public Administration Research and Theory, 20(4), 827-848.
Brodkin, E. Z., & Marston, G. (Eds.) (2013). Work and the Welfare State: Street-Level Organizations and Workfare Politics. Washington DC: Georgetown University Press.
Burawoy, M. (1998). The extended case method. Sociological Theory, 16(1), 4-33.
Denzin, N. K. (1989). The Research Act: A Theoretical Introduction to Sociological Methods (3rd ed.). Englewood Cliffs, NJ: Prentice Hall.
Diller, M. (2000). The Revolution in Welfare Administration: Rules, Discretion, and Entrepreneurial Government. New York University Law Review, 75(5), 1121-1220.
DuBois, V. (2010). The Bureaucrat and the Poor: Encounters in French Welfare Offices. Burlington, VT: Ashgate.
Edin, K., & Lein, L. (1997). Making Ends Meet: How Single Mothers Survive Welfare and Low-Wage Work. NY: Sage Publications.
Esping-Andersen, G. (1990). The Three Worlds of Welfare Capitalism. Princeton, NJ: Princeton University Press.
Greenberg, D., & Cebulla, A. (2008). The cost-effectiveness of welfare-to-work programs: A meta-anaylsis. Public Budgeting & Finance, 10(2), 112-145.
Grogger, J., & Karoly, L. A. (2005). Welfare Reform: Effects of a Decade of Change. Cambridge, MA: Harvard University Press.
Handler, J., & Hasenfeld, Y. (1991). The Moral Construction of Poverty: Welfare Reform in America. Newbury Park, CA: Sage Publications.
Handler, J., & Hollingworth, E. J. (1971). The “Deserving Poor”: A Study of Welfare Administration. Chicago: Markham Publishing.
Hasenfeld, Y. (2000). Organizational Forms as Moral Practices: The Case of Welfare Departments. Social Service Review, 74(3), 329-351.
Lipsky, M. (1980). Street-level Bureaucracy: Dilemmas of the Individual in Public Services. New York: Russell Sage Foundation.
Miller, G. (1983). Holding Clients Accountable: The Micro-Politics of Trouble in a Work Incentive Program. Social Problems, 31(2), 139-151.
Moynihan, D. P., & Herd, P. (2010). Red Tape and Democracy: How Rules Affect Citizenship Rights. American Review of Public Administration, 40(6), 654-670.
Pavetti, L., Trisi, D., & Schott, L. (2011). TANF Responded Unevenly to Increase in Need During Downturn. Washington, DC: Center on Budget and Policy Priorities.
Piven, F. F., & Cloward, R. (1971). Regulating the Poor: The Functions of Public Welfare. New York, NY: Vintage.
Weaver, R. K. (2000). Ending Welfare As We Know It. Washington, DC: Brookings.
Zatz, N. (2006). What Welfare Requires from Work. UCLA Law Review, 54(2), 373-464.
Evelyn Z. Brodkin
School of Social Service Administration
University of Chicago
969 E 60th St
Chicago IL 60637
 The author is grateful for research support for this project from the National Science Foundation (SES-0129643 and 129643), the Ford Foundation, and the Open Society Institute. Portions of this article are adapted from E. Brodkin, 2013. "Commodification, Inclusion, or What: Workfare in Everyday Organizational Life." In Work and the Welfare State: Street-Level Organizations and Workfare Politics, edited by E. Brodkin and G. Marston, 143-166. Copyright 2013 by Georgetown University Press.
 There are a variety of interesting European studies that take this approach in order to offer taxonomies of workfare programs. See, for example, Bonoli and Natali 2011, Kildal 2001, Vis 2008.
 Some of the most sophisticated quantitative analyses of US workfare suggest different outcomes for differently situated segments of the population (Greenberg and Cebulla 2008; Grogger and Karoly 2005). They can "control" for demographic differences, but cannot "control" for programmatic experience, that is, how individuals were sorted into programmatic activities (did placements respond to individual circumstances or match needs to service providers?) or the quality or content of what happened within them. That said, the aggregate findings are quite interesting. Generally, those who were among the best off (i.e., those with high school education and previous work experience) were more likely than others to leave welfare for work, at least during the economic boom years of the late 1990s. However, even then, their wage rates placed them only just above the poverty line, assuming that work was available to them on a full-time basis and their hours were consistent. A middle group of recipients appear to have moved in and out of poverty as they churned between lower-wage jobs and welfare. The third group, those who were among the most disadvantaged, were at greatest risk of ending up with no work and no welfare.
 I place my first reference to "welfare reform" in quotation marks in recognition of the problematics of US policy discourse. The national legislation that made social assistance conditional on work is colloquially referred to as welfare reform. However, the term "reform" tends to signify correction or improvement. It is more a political référentiel than a clear construct. I will use the term in an order to reclaim it from its more useage by subjecting so-called "reform" to critical scrutiny.
 Previously, under the terms of Aid to Families with Dependent Children (AFDC), the federal government provided an open-ended guarantee to states that it would reimburse part of the cost of their welfare payments. It set no time limits on the duration of benefits it guaranteed. The term entitlement refers to this guarantee.
 Among TANF's key features: it ended the federal guarantee of income support and place a lifetime 5-year limit on assistance; it imposed work requirements as a condition for receiving assistance, with adults required to participate in work activities within a maximum of two years of receiving assistance; it capped federal expenditures, providing a $16 billion block grant over five years. For a more detailed look at the legislative process and the Act's details, see Weaver (2000).
 Individual states may vary in the terms they set for extending time limits and some attach work requirements to those extensions.
 According to Esping-Anderson (1990:21-22): “De-commodification occurs when a service is rendered as a matter of right, and when a person can maintain a livelihood without reliance on the market.”
 These benchmarks substantially increased when TANF was reauthorized in the Deficit Reduction Act of 2005. Under these provisions, states must engage at least 50 percent of their assistance caseloads in a specific list of "countable" work activities for at least 30 hours a week (20 hours a week for single parents with a child under 6), or face financial penalties. There is a separate requirement that 90 percent of two-parent families be engaged in work activities for 35 hours a week (55 hours for those receiving child care subsidies).
 In debate on this question in the House, Rep. Benjamin Cardin (D-MD) argued, “If child poverty goes up, we have failed.” Rep. Nancy Johnson (R-CT) acknowledged that the amendment raised an important concern. “This is something we want to achieve, to see women and children do better in America,” she said. “But if you use poverty as a goal, you will reduce motivation for achieving other goals,” she emphasized. The Cardin Amendment was defeated in committee by a 5-8 vote.
 The study used a triangulation method to cross-check different forms of data with each other (Denzin, 1989), subjecting inconsistent findings to special scrutiny through successive, iterative analysis.
 As I have elaborated elsewhere: “...street-level practitioners will respond to an individual calculus of costs and benefits that derive from the ratio of resource availability to demand burden (R: D) as moderated by incentives. According to this calculus, one can assume that caseworkers will select action A over B when A is less costly and more rewarding. It follows that management strategies that change the informal calculus of costs and benefits will result in different patterns of discretionary choice.) Brodkin 2013a:165-166.
 For a fuller discussion of these strategies, see Lipsky 1980.
 Handler and Hasenfeld (1991) argue that "(b)ecause much greater emphasis is placed on the symbolic than on the substantive consequences of welfare policy, its implementation tends to focus on structural features designed to affirm the distinctions between deserving and undeserving poor, and to certify eligibility for welfare benefits. We term these features myths and ceremonies because their main function is to confirm the dominant cultural norms about the poor...." (11).
 See Lipsky (1980) for a general discussion of the tendency of street-level workers to redefine both their tasks and their clients. In this case, the pressures to demonstrate performance intensified after an initial period of sharp caseload decline, generally understood as moving many of the most "work-ready" off the caseload. At that point, as one manager explained, "we're left with cats and dogs now."
 Hasenfeld refers to this as the moral construction of clients (Hasenfeld 2000:337).